| 摘要 |
This application is aimed at authenticating a user on a device using not just the users password, or PIN, but also biometric information that is associated with the user, such as the way he walks or touches the device. When the user interacts with the device 02, there are sensors that generate raw sensory data 04. It then uses this data to create data points and add them to a user data set 08. An algorithm is then used to identify and classify data clusters 16. When the user tries to input his password, or sign in, this creates a new data point, which is then added to a particular cluster 24. A further classifier is then used to give the data point a confidence score, and depending upon whether or not the confidence score is above a threshold, the user will be authenticated 28, 30, 32. |
